1. Field of the Invention
this invention relates to a method and apparatus for conducting cytotoxicity assays on human biopsied tumor cells.
2. Description of the Prior Art
In 1955, Karnofsky presented to an audience of skeptics the ability of alkylating agents and antimetabolites to exhibit some chemotherapeutic activity in a limited number of human tumors. Tumor chemotherapy has greatly advanced since then and maintains an important role in the treatment of many tumors. Currently, chemotherapy can cure more than 16 tumor types, including hematological neoplasms; sarcomas; testicular, gestational, trophoblastic, and Wilm's tumors; and small cell lung and ovarian cancers. Other tumors curable in the adjuvant setting are breast and colon cancers. Advances in chemotherapeutics are ongoing: more effective drugs, drug analogs with less toxicity, and drugs modified for improved tissue uptake and longer plasma half life are being developed. DeVita, V.T. Jr, Cancer Principles and Practice of Oncology, Principles of Chemotherapy (V.T. DeVita, Jr., S. Hellman, S.A. Rosenberg eds.), J.B. Lippincott Co., Philadelphia, Pa., 1989.
Currently, tumor chemotherapeutic treatment is based upon standard practices resulting from empirical drug selection or established protocols. Von Hoff, D.D., L. Weisenthal, In Vitro Methods to Predict for Patient Response to Chemotherapy, Advances in Pharmacology and Chemotherapy, 17:133-156, 1980. Woltering, Eugene A., Administration of Cytotoxic Chemotherapeutic Agents Without Predictive Information, Laboratory of Medicine, 21:82, 1990. Variations in patient response to standard therapy are often the result of the highly heterogeneous nature of human tumors, both among different tumor types and within the same tumor type in an individual patient. This heterogeneity is reflected in the chemosensitivity of malignant cells. Variations in response can render unattainable the primary goals of chemotherapy--to maximize the effects of therapy on the tumor and to prevent side effects of therapy for the patient. Chemotherapy can offer the best chance of survival but it also has adverse effects that can be devastating, such as patient toxicity, immune system suppression, loss of time due to an ineffective treatment regimen, and the development of drug resistance.
A predictive assay is particularly important for cancer chemotherapy since is identifies ineffective drugs whose side effects are potentially life-threatening. The development of an in vitro assay which could predict the response of an individual's tumor cells to chemotherapeutics has been a longstanding objective in cancer research. The pioneers in this field include: Hamburger, Q.W., and S.E. Salmon, Primary Bioassay of Human Tumor Stem Cells, Science, 197:461-463, 1977, Hamburger, Anne W., The Human Tumor Clonogenic Assay as a Model System in Cell Biology, International Journal of Cell Cloning, 5:89-107, 1987, Scheithauer, W., G.M. Clark, S.E. Salmon, W. Dorda, R.HY. Shoemaker, D.D. Von Hoff, Model for Estimation of Clinically Achievable Plasma Concentrations for Investigational Anticancer Drugs in man, Cancer Treatment Reports, 70:1379, Shoemaker, R.H., M.K. Wolpert-DeFilippes, R.W. Makuch, Application of the Human Tumor Clonogenic Assay for New Drug Screening, Stem Cells, 1:308, 1981, Shoemaker, R.H., M.K. Wolpert-DeFilippes, R.W. Makuch, Use of the Human Tumor Clonogenic Assay for New Drug Screening, Proc. Amer. Assoc. Cancer Research, 24:1231, 1983, Shoemaker, R.H., M.K. Wolpert-DeFilippes, J.M. Venditti, IV., Human Tumors in the Screening of Cytostatics, Behring Inst. Mitt., 74:262, 1984, Alberts, D.S., H.S.G. Chen, Cloning of Human Tumor Cells, (S.E. Salmon ed.), 351-359Alan R. Liss, Inc., New York, N.Y., 1980, Alberts, D.S., H.S.G. Chen, L. Young, T.E. Moon, S.A. Loesch, E.A. Surwit, S.E. Salmon, Improved Survival for Relapsing Ovarian Cancer (OVCA) Patients Using the Human Tumor Stem Cell Assay (HTSCA) to Select Chemotherapy, Proc. Am. Assoc. Cancer Research, 22:461, 1981, Alberts, D.S., S.E. Salmon, E.A. Surwit, H.S.G. Chen, T.E. Moon, L. Young, Combination Chemotherapy (CRx) In Vitro With the Human Tumor Stem Cell Assay (HTSCA), Cancer Chemother, Pharmcol., 6:253, 1981, Von Hoff, Daniel D., James Casper, Edward Bradley, John Sandbach, Donna Jones, Robert Makuch, Association Between Human Tumor Colony-Forming Assay Results and Response of an Individual Patient's Tumor to Chemotherapy, American Journal of Medicine, 70:1027-1032, 1981., Von Hoff, Daniel D., Gary M. Clark, Brian J. Stogdill, Michael F. Sarosdy, Michael T. O'Brien, James T. Casper, Douglas E. Mattox, Carey P. Page, Anatolio B. Cruz, and John F. Sandbach, Prospective Clinical Trial of a Human Tumor Cloning System, Cancer Research, 43:1926-1931, 1983, Hanauske, Axel-R., Daniel D. Von Hoff, Clinical Correlations with the Human Tumor Cloning Assay, Cancer Investigation, 3(6):541-551, 1985., Von Hoff, Daniel D., In Vitro Predictive Testing: The Sulfonamide Era, International Journal of Cell Cloning, 5:179-190, 1987, Von Hoff, Daniel D., Commentary, He's Not Going to Talk About In Vitro Predictive Assays Again, Is He?, Journal of the National Cancer Institute, 82:96-101, 1990.
Current in vitro chemosensitivity methods include: clonging of human tumors on double layer soft agar, i.e. human tumor cloning assay (HTCA); subrenal capsule assay method, fluorescent cytoprint assay (Rotman RIVCA) and tritiated thymidine uptake assay.
The traditional in vitro method of growing human tumor cells in semi-solid agar developed by Salmon and Hamburger is referred to as the human tumor clonging assay (HTCA). In this method, solid tumors or malignant fluid from cancer patients may be used as the tumor cell source. The tumor specimens are mechanically and enzymatically dissociated to fulfill the requirement of a single-cell suspension. When short-term drug treatment is being evaluated, the single-cell suspension is incubated in media with or without the therapeutic drug. After the cells are exposed to the drug for one hour, control and treated cells are plated on agar. For continuous drug treatment, the drug is added directly to the top agar layer containing cells in a 2-layer system. In both cases, the cells are incubated for 14-21 days and observed for colony formation. The difference in the number of colonies counted in plates containing drug treated cells and in control plates is used to determine drug responsiveness, Shoemaker, Robert H., Mary K. Wolpert-DeFilippes, David H. Kern, Michael M. Lieber, Robert W. Makuch, Jeannete R. Melnick, William T. Miller, Sydney E. Salmon, Richard M. Simon, John M. Venditti and Daniel D. Von Hoff, Application of a Human Tumor Colony Forming Assay to a New Drug Sensitivity, Cancer Research, 45:2145-2153, 1985, Woltering, Eugene A., Tumor Chemosensitivity Testing: An Evolving Technique, Laboratory Medicine, 2:82-84, 1990.
The subrenal capsule assay method differs from the traditional HTCA by utilizing tumor fragments, measuring the responsiveness of multiple cell populations rather than single-cell suspensions. Subrenal capsule is an in vivo assay utilizing human tumor specimens as first-generation transplant xenografts in athymic mice. The predictability of drug resistance with the subrenal capsule assay has not been found to be superior to the HTCA. The greatest advantage is that more tumor specimens can be successfully grown using subrenal capsule than HTCA, Woltering, Eugene A., Tumor Chemosensitivity Testing: An Evolving Technique, Laboratory Medicine, 2:82-84, 1990. However, due to the expense involved in maintaining mouse colonies and the technical expertise required, the subrenal capsule assay has not become a routine test.
Rotman and coworkers have developed an in vitro chemosensitivity method (RIVCA) also referred to as the fluorescent cytoprint assay (FCA). In this assay, tumor fragments are exposed to the drug and cultured. The viability of the tumor cells is measured by their ability to hydrolyze fluoresce in diacetate and retain the fluorescein. The difference in the number of fluorescent fragments before and after drug treatment is used as a measure of drug response. RIVCA is not amenable to fluid specimens or to solid specimens yielding small cell aggregates, since only aggregates larger than 50-100 cells can be photographically recorded. The predictability of drug resistance with RIVCA is similar to that of HTCA, but a greater number of tumors can be grown in vitro and evaluated using RIVCA, Woltering, Eugene A., Tumor Chemosensitivity Testing: An Evolving Technique, Laboratory Medicine, 2:82-84, 1990.
Due to difficulties in obtaining single-cell suspensions from solid tumors, long incubation times, and poor tumor growth in the HTCA, alternative methods have been evaluated. One of these methods uses the incorporation of a radionucleotide, such as tritiated thymidine, during DNA synthesis as an indication of cell viability and proliferation. Tumor preparations are exposed to drugs, either short-term or continuously, and cultured in liquid medium. Tritiated thymidine is added and the culture incubated for 16-24 hours. Incorporated radionucleotides are harvested and counted with a scintillation counter. A decrease in the uptake of tritiated thymidine by tumor cells exposed to cancer chemotherapeutics indicates sensitivity of the tumor to the drug. This assay has several advantages over HCTA, SRA, and RIVCA: a shorter culture period (4-6 days) is required, a smaller sample size can be assayed, and the strict requirement for single-cell suspensions--a goal often unachievable for solid tumors--is eliminated. Another benefit of this assay is that the determination of tumor growth is quantitative and automated, in contrast to the subjective counting of colonies by a tissue culture technician. The clinical relevance of a radionucleotide detection system in an in vitro chemosensitivity assay has been documented: Kern, David H., Carol R. Drogemuller, Michale C. Kennedy, Susanne U. Hioldebrand-Zanki, Nobuhiko Tanigawa, and Vernon K. Sondak, Development of a Miniaturized, Improved Nucleic Acid Precursor Incorporation Assay for Chemosensitivity Testing of Human Solid Tumors, Cancer Research, 45;5435-5441, 1985; Daidone, Maria Grazia, Rosella Silvestrini, Ornella Sanfilippo, Nadia Zaffaroni, Marco Varini, Mario Delena, Reliability of an In Vitro Short-Term Assay to Predict the Drug Sensitivity of Human Breast Cancer, Cancer, 56:450-14 456, 1985, Tanigawa, Nobuhiko, David H. Kern, Yorinori Hikasa, and Donald L. Morton, Rapid Assay for Evaluating the Chemosensitivity of human Tumors in Soft Agar Culture, Cancer Research, 42:2159-2164, 1982, Wilson, A.P., C.H.J. Ford, C.E. Newman, A. Howell, A Comparison of Three Assays Used for the In Vitro Chemosensitivity Testing of Human Tumours, British Journal of Cancer, 49:57-63, 1984. One group of investigators assaying cells from a variety of solid tumor types (breast, lung, and ovarian cancers, melanomas and sarcomas) found that 80% of the specimens were evaluable (280/351) with 100% accuracy in predicting resistance and 50% accuracy in predicting sensitivity. Kern, David H., Carol R. Drogemuller, Michael C. Kennedy, Susanne U. Hildebrand-Zanki, Nobuhiko Tanigawa, and Vernon K. Sondak, Development of a Miniaturized, Improved Nucleic Acid Precursor Incorporation Assay for Chemosensitivity Testing of Human Solid Tumors, Cancer Research, 45:4335-5441, 1985. In another report, assaying only cells derived from breast cancers, the prediction of tumor sensitivity and resistance was 75% and 81% accurate, respectively. Daidone, Maria Grazia, Rosella Silvestrini, Ornella Sanfilippo, Nadia Zaffaroni, Marco Varini, Mario DeLena, Reliability of an In Vitro Short-Term Assay to Predict the Drug Sensitivity of Human Breast Cancer, Cancer, 56:450-456.
In order to conduct tumor sensitivity assays the tumor must be maintained in culture. Epithelial cells are the select culture medium in this field. In a recently published report, Von Hoff, Daniel D., Commentary, He's Not Going to Talk About In Vitro Predictive Assays Again, Is He?, Journal of the National Cancer Institute, 82:96-101, 1990, based on nearly 14,000 tumor samples, only 3,886 or 27.9% had sufficient in vitro growth for evaluation of drug sensitivity. Recent modifications of culture conditions and the development of sensitive detection methods have increased the capacity to obtain evaluable specimens. Hanauske, Axel-R., Daniel D. Von Hoff, Clinical Correlations with the Human Tumor Cloning Assay, Cancer Investigation, 3(6):541-551, 1985, Shoemaker, Robert H., Mary K. Wolpert-DeFilippes, David H. Kern, Michael M. Lieber, Robert W. Makuch, Jeannete R. Melnick, William T. Miller, Sydney E. Salmon, Richard M. Simon, John M. Venditti, and Daniel D. Von Hoff, Application of a Human Tumor Colony Forming Assay to New Drug Sensitivity, Cancer Research, 45:2145-2153, 1985. The specific problem of inadequate in vitro tumor growth has been explored extensively. To achieve adequate tumor growth, a defined, selective medium is required that allows tumor cells, most commonly of epithelial origin, to actively proliferate while inhibiting the proliferation of normal cells, such as fibroblasts. It has become apparent, however, that "traditional" growth media and high serum concentrations are not optimal for epithelial tumor cell growth. Reid, Lola M., Generic Methods for Defined Hormonal and Matrix Conditions for Studies of Growth or Gene Expression in Differentiated Epithelia, Methods in Molecular Biology, (J.W. Pollard, J.M. Walker, eds.), Volume 5: Tissue Culture. Growth medium containing a high calcium (greater than 1 mM) and high serum (10-25%) concentration enhances the proliferation of fibroblasts. In contrast, epithelial cells exhibit the best growth in a low calcium (approximately 0.4 mM) and a low serum (1% and below) environment. Additionally, serum in the medium contributes to inconsistent results between assays due to lot-to-lot variations in the concentrations of several critical components in the serum. A low serum concentration in the medium reduces the impact of this variability. However, since epithelial tumor cells require specific growth factors and hormones which are present in the serum, reduction of the serum concentration necessitates the supplementation of those growth factors and hormones. Barnes, David and Gordon Satro, Methods for Growth of Cultured Cells in Serum-Free Medium, Analytical Biochemistry, 102:255-270, 1980. Therefore, a growth medium containing low calcium and serum concentrations supplemented with defined growth factors and hormones allows preferential growth of epithelial tumor cells, resulting in an increase in the number of evaluable specimens. Reid, Lola M., Generic Methods for Defined Hormonal and Matrix Conditions for Studies of Growth or Gene Expression in Differentiated Epithelia, Methods in Molecular Biology, (J.W. Pollard, J.M. Walker, eds.), Volume 5: Tissue Culture, Crickard, Kent, Ulla Crickard, Mahmood Yoonessi, Human Ovarian Carcinoma Cells Maintained on Extracellular Matrix Versis Plastic, Cancer Research, 43:2762-2767, 1983.
Roswell Park Memorial Institute 1640 (RPMI) (Life Technologies, Grand Island, N.Y.) is a basal medium containing inorganic elements, energy sources, vitamins, amino acids, and a low concentration of calcium (0.67 mM). RPMI, however, lacks the hormones and growth factors often necessary for proliferation of epithelial tumor cells. Various hormones and growth factors are typically added individually to the growth medium according to the requirements of the cell type being grown. Barnes, David and Gordon Sato, Methods for Growth of Cultured Cells in Serum-Free Medium, Analytical Biochemistry, 102:255-270, 1980, Ham, R.G., Importance of the Basal Nutrient Medium in the Design of Hormonally Defined Media, Cold Spring Harbor Laboratory, 9:39-60, 1982.
In a review of 2300 patients, the correlation between an in vitro chemosensitivity assay and actual patient response indicated that the predictability of true sensitivity is 69% and true negative predictability 91%. Scheithauer, W., G.M. Clark, S.E. Salmon, W. Dorda, R.H. Shoemaker, D.D. Von Hoff, Model for Estimation of Clinically Achievable Plasma Concentrations for Investigational Anticancer Drugs in Man, Cancer Treatment Reports, 70:1379, Von Hoff, Daniel D., James Casper, Edward Bradley, John Sandbach, Donna Jones, Robert Makuch, Association Between Human Tumor Colony-Forming Assay Results and Response of an Individual Patient's Tumor to Chemotherapy, American Journal of Medicine, 70:1027-1032, 1981, Von Hoff, Daniel D., Gary M. Clark, Brian J. Stogdill, Michael P. Sarosdy, Michael T. O'Brien, James T. Casper, Douglas E. Mattox, Carey P. Page, Anatolio B. Cruz, and John F. Sandbach, Prospective Clinical Trial of a Human Tumor Cloning System, Cancer Research, 43:1926-1931, 1983, Hanuske, Axel-R., Daniel D. Von Hoff, Clinical Correlations With the Human Tumor Cloning Assay, Cancer Investigation, 3(6):541-551, 1985. These correlations have been the result of either retrospective or prospective single-arm studies. Only two prospective randomized trials have been performed. One prospective study was conducted with ovarian cancer patients in which the response rates, though not statistically significant, were 65% for the standard chemotherapy arm and 85% for treatment based on in vitro assay results. Welander, C.E., T.M. Morgan, H.D. Homesley, Multiple Factors Predicting Responders to Combination Chemotherapy in Patients With Ovarian Cancer, In Human Tumor Cloning (S.E. Salmon, J.M. Trent eds.) Orland: Grune and Stratton, 521-534, 1984. A larger scale prospective randomized trial was conducted on 133 advanced metastatic cancer patients. Patient response rates were 21% for those who received single-agent chemotherapy based on in vitro assay results and only 3% in patients who received a clinician's choice of a single agent. Von Hoff, Daniel D., Commentary, He's Not Going to Talk About In Vitro Predictive Assays Again, Is He?, Journal of the National Cancer Institute, 82:96-101, 1990. These studies begin to supply reliable data supporting the general use of an assay to predict patient response to chemotherapeutics.
An in vitro chemoresponse assay is not in general use because of a number of obstacles which have contributed to the lack of clinical feasibility of the assay. Currently, in vitro drug response assays are performed in university hospitals and a few specialized service centers. These institutions generally require the transportation of the specimen, resulting in a loss of often more than 24 hours before specimen processing can begin. Within this 24 hour period, specimen viability declines significantly. Additionally, an in vitro assay is not in general use because of the lack of truly effective cancer drugs and difficulties in trying to model in vivo pharmacokinetics. Other reasons why in vitro chemoresponse tests are not in general use include: technical complexity of the assays; the inability to grow tumor cells in vitro; long turn-around time; large number of tumor cells required to conduct an assay; low percentage of specimens suitable for the assay; and the lack of quality control for drugs and medium.